{"ID":2866043,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.21185","arxiv_id":"2509.21185","title":"Hybrid Real- And Complex-Valued Neural Network Concept For Low-Complexity Phase-Aware Speech Enhancement","abstract":"In this paper, we propose hybrid real- and complex-valued neural networks for speech enhancement. Real- or complex-valued models are either inefficient or present high complexity. We devise a straightforward design method for extending a real-valued network into its hybrid counterpart. Based on speech intelligibility and quality metrics, we compare the real, complex, and hybrid versions of a convolutional and a convolutional-recurrent architecture. The hybrid network consistently outperforms its counterparts with the same number of parameters. Additionally, the hybrid models' complexity in terms of multiply-accumulate operations is substantially lower than that of their counterparts.","short_abstract":"In this paper, we propose hybrid real- and complex-valued neural networks for speech enhancement. Real- or complex-valued models are either inefficient or present high complexity. We devise a straightforward design method for extending a real-valued network into its hybrid counterpart. Based on speech intelligibility a...","url_abs":"https://arxiv.org/abs/2509.21185","url_pdf":"https://arxiv.org/pdf/2509.21185v1","authors":"[\"Luan Vinícius Fiorio\",\"Alex Young\",\"Ronald M. Aarts\"]","published":"2025-09-25T14:00:57Z","proceeding":"eess.AS","tasks":"[\"eess.AS\"]","methods":"[]","has_code":false}